Darwin and Derrida

Darwin and Derrida:Cognitive Literary TheoryAs a Species of Post-Structuralism

Ellen SpolskyEnglish, Bar-Ilan

Abstract Rather than denying the insights of post-structuralist theory, literary in-

terpretation and theory with an evolutionary cognitive perspective actually nestles

nicely within a central niche of deconstructionist thinking, that is, the critique of

representation.What we learn from recent cognitive science is that the meanings of

texts are indeed unstable and dependent upon contingent contexts. While theories

of neuronal activity can be understood as analogous to the critique of representation,

the cognitive evolutionary argument supports Stanley Cavells counterproposal, that

is, that while our representational powers are not ideal they are sucient. It is pos-

sible then to argue further that the very exibility that destabilizes meaning is not

only good enough, it is responsible for our success, such as it has been, in building

and revising human cultures.

Situated within and enriching the insights of post-structuralist theory, cog-nitive literary theory conrms and claries issues previously dealt with byphilosophical, psychoanalytical, and cultural theorists and is beginning toproduce the kind of sophisticated literary scholarship rightly valued withinour profession.Were I to locate the source of my optimism in a single mo-ment of epiphany, I would refer to my having been convinced that Darwinstheory of evolution is signicantly homologous to the post-structuralist cri-tique of representation. A helpful text here is Daniel Dennetts () read-ing of Darwin as a natural theory of permanently unstable ontological cate-

Poetics Today : (Spring ). Copyright by the Porter Institute for Poetics andSemiotics.

44 Poetics Today 23:1

gories. Based on Darwins observation that evolved mechanisms might bereused for new purposes in new environments, Dennett (: ) ex-tends this natural argument against biological essentialism to a performa-tive theory of natural (and perpetual) meaning. In this theory languagemeaning is fundamentally contingent or, as Jacques Derrida (: ) de-scribed it in his deconstruction of J. L. Austins speech act theory, compro-mised by the unsaturability of context.This conrms, forme,HillisMillers(: ) claim that there is no escaping the performative or positionalpower of language as inscription. The deconstructionist project has madeit impossible to ignore how the rhetorical or tropological dimension oflanguage undermines our condence in stable, iterable, meaning (ibid.).My understanding of the usefulness of the cognitive way of talking aboutthe cultural production of human minds and brains is, thus, based on ananalogy between some elementary facts about the human evolved brainand the post-structuralist view of the situatedness ofmeaning and of its con-sequent vulnerability to the displacements and reversals that deconstruc-tionist criticism reveals.My neurons, it seems, are like me.They spend their time collecting mes-

sages from dierent (often dierently reliable and often conicting) sourcesand weighing them.The complicated decisions they make about whetheror not to continue the transmission (to re) seem not unlike mine on a nor-mal day of watching, thinking,making connections with other information,and nally acting, often on less than the best evidence. My decisions, thatis, are often, like my brains, preference judgments.1 Although an evolvedbrain cannot be said to have been built for all that I use it for, it has adapted,and for now at least its functioning is good enough to get me through theday. (Mostly.)The philosophical parallel to the brains not-after-all-so-oddly human

way of working is to be found in the critique of metaphysical idealism as ex-pounded, for example, in the work of Stanley Cavell. Cavells rereading ofLudwigWittgenstein in The Claim of Reason () moves philosophical dis-cussion toward a neurologically authentic plane by struggling to nd a wayto display, express, and understand how ordinary everyday truths indeedwork for us well enough most of the time. Cavells work enlightens and alsolightens the problems ofmiscommunication, arriving again and again at theclaim that it is possible to live an intelligent, satisfying, and even amoral lifewith the mental equipment that is the human inheritance. It is possible to

. Preference models were rst described in Jackendo as a way to display word mean-ing without fatal rigidity. Ellen Schauber and I () demonstrated how the model describesliterary genres.

Spolsky Darwin and Derrida 45

recover, in his words, from the tragically debilitating skepticism that rejectsgood enough knowledge in a vain struggle for an impossible ideal.But this is to jump too quickly to the happy ending. We need neither

neurology nor philosophy to tell us that human beings are hardly per-fect knowers. Sometimes we get it, but often we just get confused. Peopleare often rather pathetic misunderstanders not to mention megaforgetters,chronic confusers, malaproposers.We laugh at the human comedy of tele-vision sitcoms and cry at the high cultural texts that display the tragiclimits of our understanding. Homer Simpson stumbles through, but Oedi-pus,Hamlet, andKingDavidleaders ofmen, philosophers, poetsusingtheir senses and all the intuition they can muster, fail to know enough whenit matters most.From an evolutionary perspective, then, Aristotles opening assertion in

hisMetaphysics is incomplete. Allmen, by nature, he said, desire to know.Forfeiting his gnomic elegance, the sentence might be expanded thus: Itis a good guess that the survival of the human species depended and prob-ably still depends on an ability to collect, collate, and apply reasonably re-liable information about the environment to the projection of possibilities(including counterfactual possibilities), and to decision-making processes.If this assumption is true, then it is probably true that all people by naturedesire to know and that those who did not innately desire to know andwho did not develop reasonably accurate (but not necessarily perfect) waysof assessing their relationships to the world around them have long sincedied o.One general answer to the question of how the capacity to know has de-

veloped with the requisite exibility is suggested by a modular theory ofcognition. Briey, modularity hypotheses explore the implications of ourhaving developed parallel systems of knowledge acquisition. Like other ani-mals, and even some plants, humans learn about an object inmore than oneway at once, by seeing and hearing and touching it, for example.While thishas on the whole been good for survival, modularity also produces inter-modular conicts.Consider the advantages rst. One advantage of modularity is that by

having a set of dierent receptors/processors (ears and eyes, etc.) instead ofhaving a single all-purpose one, we are equipped to respond variously tothe various kinds of energy in the world (to sound waves and to light waves,to words and to smiles). Another positive consequence of modularity is thatif one system fails all is not lost. The various modules are suciently inde-pendent to defer and perhaps avoid general shutdown, often allowing theindividual to survive even if compromised.There is, however, a price to be paid for the relative insularity ofmodules,


and that is that translation systems are needed to integrate the knowledgereceived and produced in the separatemodules (Jackendo ). And herewe can make use of another evolutionary postulate: If the translation sys-tems between modules were perfectif knowledge from one module wereso well translated into knowledge from another that the two were entirelythe same (that is, seemed the same to consciousness and functioned in thesame way)then we would lose the advantage of being made aware ofthe dierences. If you know that there is a stream nearby both by noticingthe relative richness of vegetation in a linear pattern and by hearing it, youcan nd it, day or night. The modular system of knowing makes good useof a set of less than ideal receptors by combining them to produce redun-dancy, that is, conrmation, even at the cost of producing conict now andthen.The gaps or miscalibrations between the information from the dier-ent modules may be lled by analogy with memories of past experience orby inference and deduction.This story of modularity was originally told byJerry Fodor (), but the version oered here includes revisions by RayJackendo (), who rst noticed the gaps between structures of infor-mation. In Gaps in Nature: Literary Interpretation and the Modular Mind (),I adduced a further advantage of modularity: the creative potential of thegaps. Since a gap must be lled by an inference based on individual ex-perience and memory, even in similar situations dierent people will makedierent inferences.The evidence of cultural diversity (indeed the dicultyof locating and describing cultural universals) strongly suggests that the ge-netic inheritance does not predict how it will be used, nor does it control theoutcome of its processes. Assuming then that at least a gap if not a conictalways exists between what can be learned from vision and from, say, wordsor between words and touch, a cultural historian will have to understandthe historical/ideological/cultural context in order to gain insight into thequestion of why any particular intermodular conict (itself presumably anage-old physiological miscalibration; somethingHomo sapiens had long agolearned to compensate for, ignore, and even prot from), suddenly becomesa cultural crisis.2

Let me pause for a minute to punctuate. I have begun to suggest thatcognitive and evolutionary hypotheses produce new questions about his-tory and culture, questions that require further historical study. Because thehuman genetic inheritance is so complex, providing so many possibilitiesdierently actualized in dierent circumstances and by dierent people, it

. I have explored some examples of these miscalibrations in Spolsky . See also Spol-sky a, where I explore in depth a variety of creative productions that display and orderthe anxiety that results from these gaps within the specic historical context of early modernEurope.


will not be possible to answer the new and interesting questions about lit-erature and culture without consulting the historical context of the ques-tions. The evolutionary perspective will not relieve us of our responsibilityto understand how conicts produced by our inherited human brains aremodulated andmanaged within their cultural contexts.Taking this respon-sibility seriously then, it must be argued that the two kinds of studycogni-tive and culturalare noncontradictory before their complementarity canbe considered.I will argue, then, that the assumptions that emerge from the study of

evolved human brains in their successive contexts, far from being incon-sistent with post-structuralist thought, actually extend and enrich it. Myquestions and my mode of response arise of course from my own intellec-tual econiche, specically from the barrage of skeptical challenges to thehuman ability to know raised in the last forty years.The historical situationitself, in a cognitive evolutionary literary theory, provides a way to see intothe larger question of how people make or produce knowledge from thecombined resources of body (including the mind/brain) and culture. Howdoes the cognitive equipment allow people to learn from the world aroundthem? What role does culture (or do various cultures) play in that process,and how does any specic historically situated culture negotiate its claimswith the innate and grown claims of the human body and brain?What kindof knowledge can be had, and how is it acquired? How do I presume to geta piece of it and share it with you?The claim that the ultimate goal of literary theory is to tell a story about

the human mind can be traced back to Aristotle and, in modern criticism,to Northrop Frye. In his Anatomy of Criticism (), Frye claimed that the ge-neric forms of literary works have a psychological reality that is separablefrom what he dismissed as the history of taste. Literature, he assumed,displays the structure of the mind, the same claim Claude Lvi-Strauss hadmade for the material culture of Brazilian peoples in Tristes Tropiques ()and that NoamChomsky was making for syntax in Syntactic Structures ().Whatever empirical data are studied, syntax or poems, face painting or pot-tery, the goal is ideologically humanist: the proper study of humankind isthe human mind.Although the details of Fryes proposals linking the seasons to the genres

of literature have not weathered well, many cultural historians and literaryscholars still pursue the same goal and still seek to nd the theory that willdescribe the interconnections between being human and living in a culture.To what extent is artistic work predicted or projected from a culture, andhow does artistic production work to produce that culture? The paradigm Ibelieve this kind of study ts into ismost accurately called post-structuralist,


with the hyphen preserved deliberately because both parts of the word re-main important, as will become clear. The oversimplication of the phe-nomenon of post-structuralism I am about to produce here is motivated bymy intention to display as sharply as possible the multiphase emergence ofthe paradigm in the course of the twentieth century. Any more complex ex-positionwould quickly lose sight of the forest for the trees, the undergrowth,the mossy rocks, the rolling stones, the nonbiodegradable postpicnic rub-bish.As the references above suggest, the rst breakthrough was the discovery

of structure. Although there are good reasons to trace this to Ferdinand deSaussures linguistics, it was actually a discovery made simultaneously inseveral branches of the human and the physical and biological sciences aswell.3The structuralists directed our attention to connections between phe-nomena that previously had seemed unconnected, denigrating in the pro-cess the merely empirical.Methodologically itmoved social science awayfrom an interest in description and taxonomy of particulars and towardthe description of the underlying dynamic or structure, the self-sustainingand self-modifying system that describes the function of the empirical data.Scholarly work in many elds was now fundamentally revised; the task wasnot just to collect and record but to posit or discover relational hierarchiesand syntax, that is, grammars. The meaning of the structures, as Saussureargued for language, lay in the relationships, not in the words (or data orartifacts) themselves.As the structuralist perspective in language study became familiar, even

routine, and was generalized throughout the humanities and social sciences(onemight credit Umberto Ecos interpretive work on semiotics here), it be-came possible to understand several great pre-Saussurean thinkers as proto-structuralists. Among thesewereDarwin,Marx, and Freud.All of themhaddescribed substructures and superstructures, the former governing thelatter such that hitherto unexplainable or unintegrated phenomena couldbe seen as systematic.The substructures were understood by both the struc-turalists and the proto-structuralists as the equivalent of the forces of gravitythat keep the planets in their orbits. Borrowing the language of science,they also borrowed the assumption that these forces were stable and eter-nal.4 Human nature was understood to be the equivalent in the humanworld of the laws of physics.A later phase in the articulation of the paradigm was a critique of

. See Piaget for a survey of early structuralism.. This may be an instance of what Paul de Man () called the blindness that comes withinsight.While such blindness is probably inevitable, as he thought, it is not necessarily para-doxical, as he asserted, but maybe just wrong and misleading.


structuralism, now widely called deconstruction. Its roots were in philoso-phy rather than in the social sciences. It also has founders and proto-deconstructors. Aspects of human life and culture (motherhood, poverty)that had been understood to be as permanent as Platonic ideals and had,after the Romantics, been demoted to human but still stable aspectsof nature were now redescribed as indeed changeable because historicallydetermined. Nietzsche was noticed as having called attention to this inhis critique of the historical scholarship of his day.5 The historicizingof philosophical and historical categories of knowledge had already beenadumbrated in theoretical physics as the recognition of the dependence ofknowledge on the spatial location of the observer.6

The deconstruction of language, indeed of representation in general,which Derrida popularized in the late sixties and the seventies was wellunderway in the work of Husserl, Heidegger, and the later Wittgenstein.Husserl, for example, challenged both concepts (the concept of man) anddistinctions that had been assumed to be axiomatic in metaphysical phi-losophy. It was now possible to appreciate the power of the conventionalto make itself felt as natural and thence to surmise that the hierarchies ofthe binary oppositions and their weightings, which had been thought to beas natural as the orbit of the Earth, are in fact sustained in their currentconguration by dynamic tensionnot to say warby and within speciccultural settings.7Most important to literary studies, both theory and prac-tice, was the deconstruction of language or of logocentricity, as it was calledin the new vocabulary, itself begun by Saussure. The relationship betweena word and its referent had been assumed to have a transparent bond andnatural directionality: there is a solid real world of objects and a second-ary representational system that is assumed to mirror, dene, or describeit. In this view, a word (every word) has a primary, a real, a literal mean-ing, though it may also serve other uses. Although Saussures revolutionaryinsight into the arbitrary nature of the sign and the conventional natureof its functioning provided the crucial wedge by which objects and wordscould be pried apart and reinvented as signied and signier, he seems notto have fully recognized the price that would have to be paid for the discon-nection when the stabilizing power of convention itself could be (and was)deconstructed.However, in spite of that logical possibility, human discourse has not re-

. A key text was Nietzsche [].These links are summarized byGayatri ChakravortySpivak in Derrida : xxi.. Einsteins general theory of relativity in and Heisenbergs uncertainty principle in recognized this.. Culler contains a clear explanation of how these reversals are argued.


turned entropically to chaos, and as Malcolm Bradbury (: ) reassuredus, Just for the moment the instructions on a jar of instant coee remainmore or less usable. While human communication surely depends on therelative stability of word meaning and its iterability across contexts, themaintenance of the rich cultural life of human societies probably dependsas fully on our ability to trope or to distort the probable or conventionalmeaning of a word and to be understoodwhenwe do so.This catechresis, ormisuse of language, has long been recognized; its varieties were catalogedby the early rhetoricians as a set of devices. But the eect of the workof the deconstructionists in describing phenomena such as metaphor andirony has been to make clear the implications of these gures, exposing theweakness of the traditional distinctions between literary and ordinary lan-guage or literal and gural meaning.The result of several decades of post-structuralist argument has been to allow the emergence of an importantinsight: the functioning of human language depends on both its iterabilityand its instability. The combination is more than just a paradox of simulta-neous transcendence and limitation. It also allows a glimpse at how wordsthat are vulnerable in their instability are also usable for the propagationof new meanings.Saussures original observation, namely that meaning was in the rela-

tionship between words and not in the words themselves, was slowly under-stood to have destabilizing implications for the study of just about every-thing. Since the study of just about everything is conducted in words, theseinherent instabilities or ambiguities, previously described as literary phe-nomena parasitic on normal language (Austin ), were now under-stood as a general condition of language use, including language used toconduct scholarly debate. And if words are only unreliably anchored toreferents, their meanings determined by the context of other words andcultural artifacts, how can scholarship proceed?The eect of this simultaneous skepticism and liberation on the activity

of literary criticism, both Anglo-Saxon and French, is well known. If wepreviously had to be right, declared Stanley Fish (), we now only haveto be interesting. We have a new freedomfreedom to play, as Derridaspunning language exemplied, and to eschew concluding. As long as thecontext can change as easily as the weather, the life span of a scholarly con-clusion is not much longer than that of a cloud. The observation that thecenter cannot hold briey produced a comic vision for those who were,sociologically speaking, in the right places during the right years.The arguments continued, however, and the skepticism became too deep

to be borne lightly by many scholars. It is important to see how it was thatthe deconstructionist critics and their exciting and agrantly boundary-


breaking essays had backed themselves into a corner over the issue of rep-resentation.The entirely defensible assertion that language representationis not stable was transformed by a kind of rhetorical hyperbole into the in-defensible assertion that language cannot ever provide access to truth (evenif there were any). Surely because that assertion is so counterintuitive, itprovoked an eruption of professional hysteria eventually somewhat calmedby the concession that self-referentiality (meaning within a closed system),since that is all we have, works almost as well as what we thought we had:referentiality or representation.Fish, for example, soon regretted his dismissal of being right and re-

cast the situation in a way that returned some of his (and our) lost scholarlydignity. Being right has not disappeared as a standard, he argued; there stillare standards, but their ontological position has moved over a notch. Sincewe are always somewhere, always within a context, there is always a lit-eral meaning and a right interpretation (well, maybe a few). And there arealways some wrong onesfor now (Fish : ). Fishs clear responsesto the light-headedness induced by French rhetorical interpretation of Ger-man antimetaphysics produced a certain degree of containment, althoughit took the Paul de Man scandal nally to subdue the American literaryacademys enthusiasm for the unbounded transformational activity of de-constructive criticism.8

What has taken some time to establish, then, is not the error of the claimthat representational systems such as language provide no access to a realworld, only the absoluteness of that claim and, further, the interpretationsof that claim as comic or tragic. If human representational systems indeedprovided no access to unmediated reality, it would entail our rejection ofthe entire Darwinian program of evolution and adaptation. Heres why: ifit were the case that human beings (or any species for that matter) could notget some relatively reliable information about the world external to theirbodies, they could not survive for long, could not reproduce, etc. Note thatthis does not mean that the representational systems on which we dependare entirely or ideally reliable; it just means that they are reliable enoughto have ensured the survival of the species thus far.The evolutionary argument thus compromises the absoluteness of the

deconstructive claim but also, crucially, arms the gradience of the claim.

. Probably the most poignant evidence that the hyperbolic promotion of the recognitionthat words are unstable was indeed an unsustainable claimwas the inability of deMans goodfriends and colleagues to entirely erase, reread, or reinscribe the dram of anti-Semitic fascismin his early essays. When these boa-deconstructers, as Georey Hartman (see Bloom )called them, could not entirely or satisfactorily deconstruct some texts that they wanted todeconstruct, the power of trope found its limit. More analysis is still needed, not of de Mansjuvenilia but of the apologetics produced in their wake.


Precisely because the human species and its ways of knowing are evolved bythe accumulation of random mutations in interactions with changing envi-ronments rather than genetically engineered for the task of knowing, it isnot at all surprising that they are unstable. They are not purpose designedand are always vulnerable to further environmental change. It is just this in-stability, however, that provides the possibility for advantageous exibility.People, their ways of knowing, and their languages are responsive (a wordwithout the negative connotation of unreliable or unstable), that is, adapt-able within a changing environment.The only goal we can speak of withreference to adaptation is species survival, and the only thing required forthat is the survival of a certain number of individuals long enough to breedand rear ospring to the age when those ospring can breed.This does notmean that everyone has to understand everything or that understanding isa logically watertight, foolproof system. All it has to be is good enough.This argument produces two inferences: rst, that there has been, over

the course of human evolution, a curve of adaptational improvementtoward a good enough representational system and, second, that the curveeventually will atten out, that is, stop producing an ever more reliablesystem since, once it is good enough for the survival of the species, improve-ments due to random mutations will cease to be selected for. (To put it theother way, once a good enough level is attained, both the good enough andthe more successful representers can survive.) This hypothesis in itself sup-ports the claim that the representational system is indeed unstable but notthe claim that it is always paradoxical or always misleading. Furthermore,the counterpressure would be against the tendency for the representationalsystem to become increasingly more rigid: the exibility of the system hasits own advantages.The evolutionary success of the species would actuallybe compromised by an entirely rigid, that is, dependable, representationalsystem. As I argued in Gaps in Nature, the gap between the signier and thesignied is no tragedy; it builds in the exibility to allow the system to meetthe challenge of new contexts and to use oldwords in new combinations andwith newmeanings. It is true that deviations from conventional or expecteduses are risky. Attempts to communicate might fail or might not succeedunless buttressed with other communicationsfacial or hand gestures, pic-tures, paraphrases, and wordy explanations. And even then they might fail.But the prospect of never being able to adapt the representational system tonew contexts is worse from the point of view of species survival. Thus onecould hypothesize that the human representational system evolved in re-sponse to a tension between two needs, the need for good enough (reliableenough) representation and the need for a exible representational system.Evolution in that area would slow down when the lines of the two curves


intersected, and thus we live with a system that is a gradient version of thedeconstructive hypotheses: the system is not entirely stable; it is always opento catechresis, that is, to deliberate rhetorical hijacking or troping. And thatvulnerability is just what allows creative innovation, keeping the speciesgoing at the two jobs that never get done, survival and adaptation.In sum, both the deconstructionist debates of the last thirty years and the

evolutionary argument collude in stripping us of our innocence.We are nolonger able to continue as if words simply mean what they say, as if we didnot know that words cannot be entirely reliably identied with the thingsthey normally, habitually represent (even though they often do just that)or that language cannot be misread, since we now understand that itsnature, its cultural function is to be available for misreading. A misreadingin this sense is a judgment about the suitability of a reading in a context,not about any absolute or objective meaning.9 Since words cannot alwaysbe identied with what they normally represent (a matter of numericalprobability in a context), in principle the system is entirely destabilized.However, it works ne a lot of the time, although it is always at risk. Forbetter or for worse, familiar language structures may be spoken in new con-texts, may be slanted, troped, or otherwise betrayedforced, as HumptyDumpty insistedtomean what their masters want them to. Literary texts,not tomention diplomatic documents, historical records, diaries, andmanyother genres, in fact depend on this margin for their creativity.The systemis good enough for most of us to get through the day with no more thanthe accustomed undertow of misunderstanding. And often it is just what isneeded.We might have been back where we started, as indeed those who de-

clared that theory changed nothing argued.10 But we are not, because thereis another stage of post-structuralism to be reckoned withdierent anddicult to resist. This disturbance arises from the recognition of the pos-sibility, exposed in the powerful rhetoric of Michel Foucault, that if therepresentational center is indeed movable, as it is now understood to be,then it is probably manipulable. It does not just change, it is changed bysomeone or some group (Rabinow : .) In this phase of the post-structuralist debate it was repeatedly argued that theoretical hypotheses ofstructures are not only out there somewhere in the contexts of the schol-

. There is an instructive parallel here with Austins () felicity conditions. Appropriate-ness, rather than truth, is the standard by which the success of the utterance is judged. Notealso the parallel with the Darwinian idea of tness. See Spolsky b for a more detailedversion of this argument.. For the argument that theory changes nothing, see Fish : , : ; andKnapp and Michaels , .


arly world (though they are that) but are also, and in chartable ways, de-termined by the interests and contexts of their proposers and supporters.The subject of subjectivity becomes central. Lvi-Strauss () never askedwhy the unnaturalness of infanticide should be manifest by asymmetricalface painting among the Caduveo women of eastern Brazil. It seemed tohim self-explanatory on the grounds of analogy (or reduction): both wereunnatural behaviors (killing ones children and painting ones face with-out regard for its natural features). He still relied, as Derrida pointed out,on an untenable opposition between nature and culture.11 However, oncethe question of agency and subjectivity was raised, it was immediately seento have two aspects. Who decided that infanticide is unnatural? Wouldthe Caduveo women agree with the French anthropologist? Who producesand/or polices the cultural structures that determine human self-denitionand freedom of movement within the inherited structures? And inevitably,can I, or how can I, seize that power of structuration formyself ormy group?This challenge to the assumption of structural essentialism nowmeant thatsuspicious reading was inescapable, as the reader is challenged to locate theprime mover, so recently banished by the dynamic of structuration itself.Furthermore if one agrees that someone is pushing the buttons within onesown society, it begins to seem that it may be within ones power to directchange.Suspicious reading itself is not a Derridean or Foucauldian invention.

Few could have been more suspicious than Freud about the discrepanciesbetween what was said and what was meant, where the said came fromand how and why it was distorted. But for Freud, the god-in-the-machinewas a set of dark instincts thatmight be unmaskable and understandable butprobably unappeasable. Heidegger also, it would seem, considered indi-viduals to be helpless: caught in a hermeneutic circle with nowhere to standfrom which to survey all the possibilities and no way to control the leversthat move them. For some prominent post-structuralists, Paul de Man, forexample, the source of this dark, even classically tragic situation in whichthe possibility of honest representation is warped by forces greater thanany individual is itself a mystery. It might well be objected, however, thatthe history of Europe in the twentieth century, in which the promise of sci-entic progress was mocked by the violent uses to which science was put,backlights painfully, and not so mysteriously, Foucaults recognition thatsomeones interests are served by the denition and manipulation of the

. A sustained critique of a Lvi-Strauss text is Derridas talk, entitled Structure, Sign, andPlay in the Discourse of the Human Sciences, in to the symposium at Johns HopkinsUniversity. It was translated and reprinted in Derrida .


material and cultural substructure. One of Foucaults early arguments wasthat the use of language and grammar as a metaphor for human struc-turation, a metaphor that the early structural linguists had claimed was nometaphor but truth, had concealed as much as it had revealed.War, heproposed, was the more instructive metaphor. It would not conceal thatsomeone or some class of people always benets from an established struc-ture and furthermore does so specically by insisting on its naturalness andpermanence. If mystication or mythication is insucient to maintain thestatus quo, then force can be used to do so, revealing to a suspicious readerthat the structure is not entirely natural after all.In todays usage post-structuralism (or more generally post-modernism)

is the cover-all term for the generalized suspiciousness of interpretation. Inspite of its rhetoric, however, post-structuralismdoes not replace structural-ism; post-structuralists still understand the phenomena of human bodies,minds, cultures, and theories to be structured.They may not be structuredentirely naturally and are certainly not structured entirely permanently,as was assumed at one time.They are, it is now said, constructed (and vari-ously so) by the interface of our genetic inheritance with the environmentinto which we are born, that is, by the constantly changing interaction ofindividual needs, hegemonic cultures, and an unstable class of culturallyempowered arbiters (Oyama ).The exibility of the cultural system asa whole does not mean that it does not exhibit fairly reliably repeated se-quences of events. It is not necessary to adopt the view that the world is aBorgesian encyclopedia, even though postmodernist art and literature pro-duce both tragic and comic views of a world freed frommany of the specicstructures so long assumed to be inevitable. But if I remain a structural-ist, I am also a post-structuralist because I believe both that structures aredescribable simultaneously in more than one way and that they are per-manently open to revision. I am thus a skeptic in the sense that I do notbelieve in the singleness of truth,12 and I am a suspicious reader. I feel un-settled until I can determine the assumptions of the current context andwho institutes and enforces its prevailing rules. This suspiciousness of thecontexts others have establishedagendas that they have declared or notdeclaredcombined with the distrust of rational argument as just one ofmany possibilities hasmade it attractive to simply declare ones own agendamore loudly or, in our discipline, more widely and interestingly (StanleyFish again). But the sword cuts two ways; it must be acknowledged thatothers may nd their own concerns more compelling.

. The argument for themultiplicity of truths from the scientic point of view ismade nicelyin Arbib and Hesse .


Yet if the ground for truth is no longer what it was, and if there is noother, then an oddly Buddhist resignation is produced from a supposedlyradical critique. Foucault, it seems to many, saves the day, producing a neo-Marxism usable for literary studies.Whatever is, he urges, might be other-wise. And so Foucault provides the rationale for readers and interpreterswho want to see their scholarship as praxis, as eecting their world. AfterFoucault it has become dicult to pretend that one does not know thatsocial structures produce gains and losses and not randomly. It seems to methat it would be dicult to justify not asking how the structures we investi-gate as literary or cultural historians are constructed and valued.Many feelthat traditional scholarship is newly energized by the possibility that themechanics of social structures, that is, their politics, might be understoodand that individuals or groups might be in a position to discover tools withwhich to challenge them.The possibility that discourse, for example, lit-erature, might be one of those tools understandably has been exhilaratingto scholars in a eld long patronized as decorative.My defense of the cognitive study of literature, then, having located it

within the post-structuralist paradigm, needs now to make clearer whathas already been broadly suggested, and that is how Darwin also ts there.Just as Freud and Marx provided literary scholars with productive ques-tions about their texts and their interpretive procedures, so Darwins texts,through the readings of them by recent scholars in anthropology, biology,neurology, philosophy, psychology, and literary study, have opened newways of talking about our subject.This perspective is an important new toolfor the literary scholar because it asks new questions about the relationshipbetween the biological and the cultural, between the living human bodyand its environment.The general shape of my claim is that nothing could be more adapta-

tionist, more Darwinian than deconstruction and post-structuralism, sinceboth understand structurationthe production of structures (and this is thesame thing as the production of theories of structures ad innitum)as anactivity that happens within and in response to a specic environment. It isan activity that is always already designed for cultural use but also alwaysready to be reused or redesigned as needed. It is important, however, toemphasize here that this is not a Panglossian vision in which satisfactionis always available. In fact the opposite is more likely true. Since the cul-tural/biological nexus is always in motion, it never exactly ts. It is alwayson a journey between novelty and obsolescence. The motto of this worldis certainly not whatever is, is right but more accurately, whatever tswell enough will do for now. The categories of the world, and the struc-tures of categories, remain the same or revise themselves depending on their


interrelation with other categories in their environments, but only slowly.There are no absolute unchanging categories or structures. Like the recip-rocal mutations between parasites and hosts, recategorization is constantlyin process (Dawkins ; Dennett ). The variations and revisions forboth Darwin and post-structuralists are neither divinely nor benignly di-rected. Here Darwin and Foucault must part company. Darwin would nomore have attributed change to a malignant intention, as Foucault alwaysseemed to, than to an angelic intention.Darwinism is appealing as a theory of mind and of meaning because it

is a theory of survival that depends upon adaptation (troping, reinterpre-tation, rerepresentation) by recategorization. To put it another way, it is atheory that justies the centrality of potential recategorization by describ-ing it as a mechanism for survival. It is a theory of how living organismssurvive in an unreliable environment by dynamic metamorphosis. In its ex-tensions into the realms of culture it suggests how metamorphoses spreadthroughout populations and become entrenched (Dawkins ; Sperber). Water-dwelling creatures became amphibious as the swamps driedup; Syrinx was changed into a reed to escape Pan.The comparison with Ovid is not as far-fetched as it may at rst seem

because, although the word adaptation sounds good-naturedly cooperative,Darwinism is in fact also a theory of unpredictable death and catastrophicvariation and recategorization, or as Tennyson put it, it is a theory thatunderstands nature to be red in tooth and claw. The grotesqueries withwhich Ovids stories often end bear comparison with the random variationand sudden loss that are necessary conditions of evolution under conditionsof natural selection. Ovid was often circumspect about the causes of themetamorphoses he described, which were mostly overdetermined. If theywere punishments, it may not be clear whowas the punisher. Did Syrinx de-cide she would rather be a reed than a victim of Pans lust? Or was someonepunishing her by her recategorization? Darwin similarly never credited theindividual mutant animal or plant with solving a problem of environmentalchange by deciding to develop lungs or chlorophyll.Indeed, I see the value of Darwins theory as a description and not as an

explanation of change, adaptation, and recategorization. On these groundsit is attractive to literary theory because the processes it hypothesizes for thenatural world of plants and animals, that is, spontaneous change/variation,followed by survival and loss and temporarily stable subspeciation, are con-sistent with many of the most interesting recent theories of mind, knowl-edge, meaning, and interpretation. Insofar as it can be argued that an evo-lutionary theory of how living creatures in the natural world adapt andsurvive is also a theory of mind, that is, a theory of the way the human


mind/brain adapts and learns (I am assuming these are not two dierentthings), then both theories are strengthened.Theorists working in several elds of human sciences indeed have de-

scribed the activities of minds in ways that seem parallel to Darwins de-scription of natural evolutionary processes. The case has been explicitlyargued of course for connectionist or parallel processing models of mind,but Darwinism is implicit as well in the Chomskian hypothesis of an in-nate language module that is modied in interaction with the environmentto produce knowledge of a specic language.13Wittgensteins () modelof language games as conventions similarly suggests the simultaneous sys-tematicity and plasticity that allow both meaning and meaning change.Stephen Greenblatts () view of the circulation of social energy in adynamic of challenge and containment, my discussion of genre change inGaps in Nature, and Lorraine Codes () feminist, relational epistemologyare also models of dynamic and interactive adaptation and self-regulation.Susan Oyama () makes clear the importance of what she calls con-structivist interactionism as a replacement for the misleading distinctionbetween a presumably unchangeable nature and the exibility of culture.All of these theories are Darwinian, I would claim, for at least this rea-

son: they all manage to account for systematicity, that is, for stability andpredictability, while allowing the possibility of adaptive change. Cruciallythey do so without the notion of an unchanging anchoring center, a set ofplatonic universals or literal meanings.The givens of these systems are onlyas mysterious as the architecture of the mind/brain itself (although that isstill pretty mysterious). The well-dened species of Darwinian theory islike the literal meaning of a word. Both are, at least for now, the most prob-ablemeanings of theword in a given community. Both are liable, even likelyto change eventually because they are embedded within unstable seman-tic and ecological systems. As Dennett describes Darwins description ofthe origin of species, the process begins and ends with well-dened species,but for the stages in between the dierences are innitesimally small. Thissounds a lot like the kind of dierences that poets can risk in making use ofwords for new purposes on the assumption that they want to be both origi-nal and understood. Experienced readers of poetry (or any unconventionaltext) have learned a set of cognitive procedures whereby they can makesense of novelty (Culler ; Schauber and Spolsky ). Eventually whatwas novel (metaphorical, say, using the word broadcast for radio transmis-sion instead of for sowing seed) may become probable or literal meaning,

. Dennett (: ) discusses Chomskys denial of this. Deacon () describes how thelanguage module might have evolved not before but within a context in which human lan-guage developed.


and a mutant may come to be recognized as a well-formed species. Dar-win, according to Dennett (: ), declines to play the traditional gameof declaring what the essential dierence is. 14 Remember here Wittgen-steins example of games: no single condition is required of all members ofthe category.

Well-dened species certainly do existit is the purpose of Darwins book to

explain their originbut he discourages us from trying to nd a principled

denition of the concept of a species. Varieties, Darwin keeps insisting, are just

incipient species, and what normally turns two varieties into two species is

not the presence of something (a new essence for each group, for instance) butthe absence of something: the intermediate cases, which used to be therewhichwere necessary stepping-stones, you might saybut have eventually gone ex-

tinct, leaving two groups that are in fact reproductively isolated aswell as dierentin their characteristics. (Dennett : )

So if the well-dened species is the literal meaning, it is as fuzzy a cate-gory as literal meaning ever was and as unstable, measurable by its dier-ence from other species. A permanently literal meaning can no more existthan can a natural category that will never change.The potential for changeis all that is permanent, and the direction of the change is not predeter-mined. Just as the algorithmic process of evolutionary change contributesto the survival of life on earth in changing circumstances, the systematicexibility of language keeps it able to serve changing communicative needs.Neither system changes without lurches and loss; both are self-stabilizingover time, but neither is rigid. If the systems were rigid, neither could serveits purpose.The analogy betweenDarwins stable categories and literal meaning can

be extended further. In both there is a dierence between thematerial situa-tion of continuous change, disappearance, and survival, clearly a gradientprocess, and the description of the resulting state of aairs. Naming andcategorizing, like the rest of language, have an inevitable ad hoc quality, yetonce they become entrenched, changes are not made easily. At any givenmoment, the set of names and categories available for use by any individualis only a near approximation of the set of material phenomena that mightneed describing. Speakers are bound to use words and language in a roughway to remain within a communicative community. For a small child, doggiewill do ne for all varieties of dogs, while at the same age cat will certainlynot encompass lion and tiger but may include a toy cat. Biologists of course

. Dennett (: ) notes a standard way of marking species dierentiation, the exis-tence of interbreeding, but then shows examples of exceptions to this rule. It is thus not anecessary rule.


will try to come closer to cutting the world at its joints with their termi-nology, but they also recognize that new empirical evidence (the so-calledmissing links) may some day prompt recategorization. Poets similarly try toget it right, to use the full range of language resources to make the descrip-tion t the speakers singular perception as closely as possible, although bythe communitys conventional standards the utterance may sound odd orunusualdeviant, as structuralists called it, since, and to the extent that theresult is less than conventional, it will be less easily understood.This exi-bility in categorization, even with its limitations, is extremely fortunate: itallows innovation.We can invent words when the need arises (gridlock) andcan make sense of someone elses neologisms (pied beauty).We can also in-uence our niche, that is, eect a change. A speaker can comfort his or herbeloved before a separation by a comparison to a pair of compasses, andan essayist can stimulate the political will of his or her peers by inventingthe word phallocracy.The project of cognitive literary studies is only just beginning. It will, I

hope, continue to explore the new questions that emerge from a consider-ation of literary issues in the light of various kinds of cognitive evidenceand to reconsider old issues with new evidence. Elaine Scarry (), forexample, proposes several ways literary texts take advantage of the brainsability to reproduce and understand what writers want us to envision. Myrecent study of early modern texts and pictures charts some of the ways inwhich creative worksmay provide satisfaction in a violently changing socialworld (Spolsky a). Mary Crane (), in her recent study of Shake-speare, argues for the importance of a consideration of embodied brainprocessing to an understanding of the author function in a literary text.Alan Richardsons () exploration of the growth of Romantic-era brainscience extends and deepens the ways in which post-structuralist culturalstudies can be enriched froma cognitive perspective.My assumption is that,with due precaution 15 and always taking care not to confuse the analogicalor metaphoric use of data from cognitive science with its analytic use, cog-nitive literary study is uniquely positioned to carry forward the advancesin understanding made by the post-structuralist critique of representation,to understand, that is, the simultaneous good-enoughness and the insta-bility of meaning. It will do this, I believe, without the high unseriousnessof post-structuralism.

. Literary scholars can inoculate themselves against the nave overestimation of what socialscience or evolutionary biology can oer by remembering to ask themselves: What is theprobability that their eld (as compared with mine) is not riven by competitive hypotheses?What is the probability that, while I struggle to deal with apparently irreconcilable complexi-ties, they know exactly what theyre doing, so that I may borrow their theories and empirical


References

Arbib, Michael A., and Mary B. Hesse The Construction of Reality (Cambridge: Cambridge University Press).

Austin, J. L. How to Do Things with Words (Oxford: Oxford University Press).

Bloom, Harold, Paul de Man, Georey H. Hartman, and J. Hillis Miller Deconstruction and Criticism (New York: Continuum).

Bradbury, Malcolm My Strange Quest for Mensonge: Structuralisms Hidden Hero (London: Arena).

Cavell, Stanley The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy (Oxford: Oxford Uni-versity Press).

Chomsky, Noam Syntactic Structures (The Hague: Mouton).

Code, LorraineWhat Can She Know? Feminist Theory and the Construction of Knowledge (Ithaca,NY:CornellUniversity Press).

Crane, Mary Thomas Shakespeares Brain: Reading with Cognitive Theory (Princeton, NJ: Princeton UniversityPress).

Culler, Jonathan Structuralist Poetics: Structuralism, Linguistics, and the Study of Literature (Ithaca, NY: Cor-nell University Press).

On Deconstruction (Ithaca, NY: Cornell University Press).Dawkins, Richard The Selsh Gene (Oxford: Oxford University Press). The Extended Phenotype: The Long Reach of the Gene (Oxford: Oxford University Press).

De Man, Paul Blindness and Insight: Essays in the Rhetoric of Contemporary Criticism (New York: OxfordUniversity Press).

Deacon, Terrence W.The Symbolic Species: The Co-evolution of Language and the Brain (NewYork:W.W.Norton).

Dennett, Daniel Darwins Dangerous Idea (New York: Simon and Schuster).

Derrida, Jacques Structure, Sign, and Play in the Discourse of the Human Sciences, in The Structural-

ist Controversy: The Languages of Criticism and the Sciences of Man, edited byRichardMackseyand Eugenio Donato, (Baltimore, MD: Johns Hopkins University Press).

Of Grammatolo, translated by Gayatri Chakravorty Spivak (Baltimore, MD: JohnsHopkins University Press).

Signature Event Context, Glyph : .Fish, Stanley A. IsThere aText inThis Class? The Authority of Interpretive Communities (Cambridge:HarvardUniversity Press).

DoingWhat Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and LegalStudies (Durham, NC: Duke University Press).

Fodor, Jerry A. The Modularity of Mind: An Essay on Faculty Psycholo (Cambridge: MIT Press).

data as unimpeachable evidence to resolve my controversies? Indeed the eld of evolution-ary biology is in the throes of several dierent controversies, which literary scholars are notprofessionally trained to evaluate.


Frye, Northrop Anatomy of Criticism: Four Essays (Princeton, NJ: Princeton University Press).

Greenblatt, Stephen Shakespearean Negotiations: The Circulation of Social Ener in Renaissance England (Berkeley:University of California Press).

Jackendo, Ray Semantics and Cognition (Cambridge: MIT Press). Consciousness and the Computational Mind (Cambridge: MIT Press).

Knapp, Steven, and Walter Benn Michaels Against Theory, Critical Inquiry : . Against Theory : Hermeneutics and Deconstruction, Critical Inquiry : .

Lvi-Strauss, Claude Tristes Tropiques (Paris: Librarie Plon).

Miller, J. Hillis The Triumph of Theory, the Resistance to Reading, and the Question of the Ma-terial Base, PMLA (): .

Nietzsche, Friedrich [] On the Advantage and Disadvantage of History for Life, translated by Peter Preuss(Indianapolis, IN: Hackett).

Oyama, Susan Evolutions Eye: A Systems View of the Biolo-Culture Divide (Durham, NC: Duke Uni-versity Press).

Piaget, Jean Le Structuralisme (Paris: Presses Universitaires de France).

Rabinow, Paul, ed. The Foucault Reader (New York: Pantheon).

Richardson, AlanBritish Romanticism and the Science of theMind (Cambridge:CambridgeUniversity Press).

Scarry, Elaine Dreaming by the Book (New York: Farrar Straus Giroux).

Schauber, Ellen, and Ellen Spolsky The Bounds of Interpretation: Linguistic Theory and Literary Text (Stanford, CA: StanfordUniversity Press).

Sperber, Daniel Explaining Culture: A Naturalistic Approach (Oxford: Blackwell).

Spolsky, Ellen Gaps in Nature: Literary Interpretation and the Modular Mind (Albany: State University ofNew York Press).

Elaborated Knowledge: Reading Kinesis in Pictures, Poetics Today : .a Satisfying Skepticism: Embodied Knowledge in the EarlyModernWorld (Aldershot: Ashgate).b Why and How to Take the Fruit and Leave the Cha, SubStance : .

Wittgenstein, Ludwig Philosophical Investigations, translated by G. E. M. Anscombe (Oxford: Blackwell).

Darwin and Derrida

Documents

natural theory

theory languagemeaning

cognitive literary theoryas

poststructuralist view

cognitive way

poststructuralist critique

austins speech act theory

natural argument